1. 2017;20;E85-E98Modified Percutaneous Lumbar Foraminoplasty and Percutaneous Endoscopic Lumbar Discectomy: Instrument Design, Technique Notes, and 5 Years Follow-up
   Cohort Assessment
   Hong-liang Zhao, MD, Ke-ran Song, MD, Wei-lin Shang, MD, Shu-xun Hou, MD, and Zhen-zhou Li, MD.

BACKGROUND: Conventional percutaneous endoscopic lumbar discectomy (PELD) with an “inside-outside” technique has 4.3% – 10.3% surgical failure rate, especially in central herniated discs (HDs), migrated HDs, and axillary type HDs. PELD with foraminoplasty has been used for complex HDs. Percutaneous lumbar foraminoplasty (PLF), which is performed with a trephine or bone reamer introduced over a guidewire without a protective working cannula in the original Tessys technique, can quickly cut the hypertrophied bony structure under fluoroscopic guidance, and risk injury to the exiting and traversing nerve roots.

STUDY DESIGN: A prospective cohort study.

SETTING: Hospital and outpatient surgical center.

OBJECTIVE: To evaluate the outcome and safety of modified PLF-PELD with a specially designed instrument for complex uncontained lumbar HDs.

METHOD: From April of 2007 to April of 2009, 148 patients with uncontained lumbar HDs were treated with modified PLF-PELD. Magnetic resonance imaging (MRI) checkup was performed the next morning after the operation. Outcomes of symptoms were evaluated by follow-up interviews at 3 months, 6 months, one year, and 5 years after surgery. Low back pain and leg pain were measured by visual analog scale (VAS) score (1 – 100). Functional outcomes were assessed by using the Oswestry Disability Index (ODI) and modified MacNab criteria.

RESULTS: Follow-up data were obtained from 134 cases, including 14 cases on L3-4, 78 cases on L4-5, and 42 cases on L5-S1. One hundred-eight cases were prolapse type, while 26 cases were sequestration type. Pre-operative symptoms and deficits included nerve root dermatome hypoesthesia in 98 patients (73%), nerve root myotome muscle weakness in 32 patients (23%), and weakening or disappearance of tendon reflex in 43 patients (32%). No case required conversion to an open procedure during the surgery. Low back pain and leg pain were significantly relieved immediately after surgery in all patients. MRI examination showed adequate removal of HD in all patients. VAS scores and ODI values were significantly lower at all time points after surgery than before surgery. The percentage of pain relief in leg pain was significantly higher than that in low back pain (P < 0.01). But there was no significant correlation between duration of the preoperative symptoms and the percentage of pain relief. MacNab scores at 5 years after surgery were obtained from 134 patients. Seventy-five cases were rated “excellent”; 49 were rated “good,” Five patients experienced heavier low back pain, thus being classified as “fair.” Five cases with recurrence were rated “poor.” Preoperative and postoperative (5 years follow-up) related nerve root function status was compared. Sensation and muscle strength recovered significantly (P < 0.01), while tendon reflex was not changed (P = 0.782). No patients had infections. Five patients were complicated with dysesthesia in distribution of the exiting nerve that was all operated at L5-S1. Complaints were reduced one week after treatment with medium frequency pulse electrotherapy. Five cases required a revision surgery after recurrence.

LIMITATIONS: This is an observational clinical case series study without comparison.